The Human Microbiome consists of all the organisms that comprise the many distinct microbial communities found on and in the human body. Early individual studies and the concerted initial steps of the Human Microbiome Project (HMP) have begun to define the extent of the diversity within these communities and the importance of their structure and temporal variation on human health. To establish the necessary foundation to measure the variation of communities and the functional consequences, we must: i) establish baseline measurements of these communities, including cataloging both the presence and prevalence of specific organisms and their gene content;ii) determine the physiological capabilities of each organism in the community by elaborating their gene content, to be able to interpret the functional significance of changes in community structure;iii) continue to develop new methods to characterize microbial communities and the organisms therein;and, iv) promote research in human metagenomics by engaging the research community and creating publicly available data and resources. We will take advantage of the explosive advances in new sequencing technologies to rapidly create a catalog of over 400 reference microbial genome sequences, including those of bacteria, eukaryotes, and viruses, as well as generate extensive community profiles from a number of human body sites with a focus on the oral and vaginal microbiome. We will also aggressively develop and apply new methods and technologies to further reduce the cost and expand the reach of the HMP. These data will be rapidly released to the public and, along with those of our collaborating partners in the HMP network, will define the Human Microbiome communities and provide baseline measurements for a wealth of subsequent experimental research. Public Health Relevance: The ultimate goal of the HMP is to understand how microbial community structure and function determine normal human health, development, and disease, and can contribute to new diagnostic or therapeutic tools. To do this we must first generate large, public data sets that define the organisms commonly found in healthy human subjects, and measure the extent to which they vary and the reasons for this.